Edison Conde Perez dos Santos
COPPE/UFRJ, Federal University of Rio de Janeiro, Brazil
E-mail: edison.conde@hotmail.com
Carlos Alberto Nunes Cosenza
COPPE/UFRJ, Federal University of Rio de Janeiro, Brazil
E-mail: cosenzacoppe@gmail.com
José Carlos Cesar Amorim
Military Institute Engineering (IME), Brazil
E-mail: jcamorim@ime.eb.br
Submission: 09/03/2017
Accept: 27/03/2017
ABSTRACT
This
study involves an assessment of various artificial intelligence-related
techniques which aim to produce a more robust system for sediment transport
modeling. The intelligent systems developed in this research are directly
applicable to academic knowledge and use data from a report on "water
circulation assessment in the “Linguado”
Channel and Babitonga Bay - Santa Catarina, Brazil", developed by the
Brazilian Military Engineering Institute (Instituto
Militar de Engenharia - IME). The solution employed for sediment transport
was built using an intelligent system from the conception of two hybrid models.
The first was a Neuro-Fuzzy (ANFIS) hybrid model for the study of hydrodynamic
behavior, aiming to determine flow rate in the channel. The second was a fuzzy
genetic model, able to assess sediment transport in the “Linguado” Channel. The study's conclusion compares the different
effects involved in the dredging equilibrium in the “Linguado” Channel according to this hybrid model with the results
obtained using a finite element model in the MIKE21® software.
Keywords: hydrodynamics, sediment transport, fuzzy logic, genetic
algorithm, neural network, ANFIS.
1. INTRODUCTION
Based on questioning the use of natural resources,
this paper presents a solution to assess transport sediment in water bodies
allowing a consistent assessment of the exploitation of natural resources with
minimal environmental impacts. The hybrid model proposed combines a Neuro-Fuzzy
(ANFIS) network for flow rate calculation with subsequent use of a Fuzzy
Genetic algorithm to calculate sediment transport.
Researchers such as Feigenbaum, from the University of Stanford, managed to build a
specialized system with 450 rules, using Heuristic Programming. This model,
known as HPP, is based on human knowledge and managed to perform patient
diagnoses. In addition to this model, other artificial intelligence models
based on scientific knowledge-supported rules have also been developed (RUSSEL
S., 2008).
The main advantage of mathematical
modeling is the capacity to make predictions by simulating future scenarios,
such as the presence of yet inbuilt structures or the occurrence of extreme
environmental conditions. Hydrodynamic modeling has been seeing wide
application in diagnosis with scarce available monitoring data. Hydrodynamic
modeling may be also considered as a prerequisite for sediment transport
modeling and for water quality modeling (VIEIRA, 2008).
This
study proposes a neuro-fuzzy hydrodynamic model to predict the flow rate in the
Babitonga Bay, in the Brazilian State of Santa Catarina, combined with a fuzzy
genetic model aiming to study sediment transport in the “Linguado” Channel. This analysis is necessary due to the sediment
input coming from Babitonga Bay through the Channel.
2. THEORETICAL FOUNDATION
The estuarine environment
undergoes both natural processes and human intervention. Circulating waters
suffer the influence from density differences and, mainly, from tidal movements
within the estuary, and are also affected by the local morphology. This creates
barothropic and baroclinic pressure gradients, which act on the movement and
mixing of coastal and river waters (GRACEA et
al, 2008).
The Navier
Stokes equation may be used to solve problems involving the specific case
of Shallow Waters (SW), as well as in studies about wave movement and water
circulation ("x" and "y") over a time "t". The
equation can be written as follows (BRATT et
al, 2010).
(1)
Where
“η” is water level height,” ” is the product
of gravitational force and height, and "k" is the friction
coefficient of the bottom. Some studies using Artificial Neural Networks (ANNs)
have been built based on hydrodynamic models in an attempt to overcome the
problem of a non-linear relationship between physical systems and phenomena
prediction in marine environments. Vaziri (1997) proposed an ANN model to
predict water movement in the Caspian Sea. This model attempted to indicate the
monthly level of surface waters.
Deo and
Chaudhari (1998) developed a model using ANN techniques, training algorithms
whose back propagation error maintained a cascade correlation and adjusted the
multivariable function through conjugated gradients thus predicting tides. Tsai
& Lee (1999) examined the applicability of a back propagation network (BPN)
and neuro-fuzzy (ANFIS) networks in order to predict tide change times.
Regarding
sediment transport, published studies provide some models for the transport of
pollutants based on equations for SW in an open channel. In such cases,
particle trajectories are defined and calculations are performed for the
different times of particles' movement in the current, and the model also takes
into account distributions according to turbulence regions, providing different
speed profiles over a time "t" (HINWOOD, 1979).
(2)
In the equation above, "c" is local sediment concentration;
"w" is sediment particles' fall velocity, and “εS“ is the
mixture's sedimentation coefficient. This type of model is applicable to the
transport of inert sediments in water. The present paper proposes fuzzy genetic
relations for sediment transport, with rules developed for an open channel. An
objective function has been coded, forming the binary input vectors.
3. ENVIRONMENTAL DATA
The information used to
solve the proposed model were obtained in the study conducted by Military
Engineering Institute (IME), and also through existing specific studies of
turbidity in the literature, which provide measurements of this variable for
eight points considered relevant for this study, points M1 to M8, located
throughout the Babitonga Bay, as shown in Figure 1.
Data from the sample used in this study were obtained
within 20-minute intervals and stored in the equipment itself. Data collection
instruments were placed in the various sampling points mentioned above. The
main tidal harmonic constants were estimated using Franco (1988)'s method.
Tidal form factor F was calculated using FEMAR
(2000)'s definition, which allowed the development of a tide classification.
Tide levels shown for tide stations M1 and M2 are arbitrary, while stations M3,
M4, M5, M6, M7 and M8 were leveled in relation to zero, which is represented by
the historic average of the spring ebb tides (IBGE's zero level), and to the
estimation of river contributions. Salinity and water temperature within the
area of interest of this study showed a similar behavior to general water level
variations, indicating a possible direct relation of these parameters as
functions of tide (SCHETTINI, 1999).
Table
1: Measurement stations used for the sediment transport model
Station |
Location |
Latitude |
Longitude |
Tide
station 1-M1 |
Monobóia |
260
13,80’ |
0480
25,05’ |
Tide
station 2-M2 |
Penha |
260
46,50’ |
0480
38,50’ |
Tide
station 3-M3 |
Capri
Beach |
260
10,90’ |
0480
34,00’ |
Tide
station 4-M4 |
Palmital
River (upstream) |
260
08,00’ |
0480
48,50’ |
Tide
station 5-M5 |
Joinville
Yatch Club |
260
17,50’ |
0480
46,00’ |
Tide
station 6-M6 |
“Linguado”
Channel (North) |
260
22,00’ |
0480
40,50’ |
Tide
station 7-M7 |
Remédios
Island |
260
27,40’ |
0480
34,85’ |
Tide
station 8-M8 |
“Linguado”
Channel (South) |
260
22,00’ |
0480
39,00’ |
Source: IME, 2003.
Figure 1: Tide
measurement stations used for the sediment transport model Source: IME, 2003.
4. THE HYBRID MODEL
Development of the model's architecture considered
four input variables and the operations performed on the inputs of each layer
in the ANFIS network. Network structure was formulated according to previously
defined rules, considering that the nodes in the first layer, relative to the
process of calculating input membership, relate each input of a fuzzy set with
its weight (JANG, 1993).
LAYER 1: composed by Inputs, represented by the values
observed for the variables bathymetry, tide, winds and roughness, respectively
xi, xj, xk, xl in the membership
function TSK, and Output (O1), representing the Value of the
(gaussian) membership function whose equation is shown below. The set {wi}
represents ANFIS' linear parameters.
Model variables are thus fuzzified within the respected values attributed to
them, as seen below.
LAYER 2: the
input is represented by O1, and the Output by O2, the
result of the fuzzy operation to be performed, which consists in multiplying
the membership degrees whose corresponding linguistic labels (fuzzy sets) are
to be combined. This layer's output represents the degree of activation of the
developed rule.
LAYER 3:
contains the Input set O2 and Output set O3. This stage
comprises calculation of the normalized degree of activation by applying the
model.
LAYER 4: has
Input set O3 and Output set O4. In this layer, the output
of layer “3” is multiplied by the function fi.
This function is the result of a linear combination of the values of layer 1
inputs x with the y input values of layer X, as seen
below:
O4 = w f = v1 =wi. Xi+
wj .Xj+ wk. Xk + wl .Xl+
b
(3)
LAYER 5: Input O4; Output O5.
This layer comprises calculation of the system's general output, which consists
in the sum of the qualified node outputs from layer 4. The nodes from the
second layer, represented as “Π“ product
operators, relate the antecedent connective e
(where 'e' represents an operator).
Node "N" represents the learning normalization process and the
summation node”∑” represents the
"average" operator. The inference function (FIS) for ANFIS networks
has an activation function (O4) represented by equation 3(JANG,
1993).
The
proposed model for sediment transport was built taking into account current
fields relative to the various scenarios. To determine the sediment fractions
in the riverbed, the domain of the Remobilized Fold Strip includes a single
geomorphological region, including the “Serra
do Mar” Escarpments and Reverse Faults. This region holds the
geomorphological units “Serra do Mar”
and “São Bento do Sul” Plateau, which
occur at the center and west of the research area, respectively.
The set of all fuzzy
genetic rules and their relation with the chromosomes defined the search space.
Each chromosome was constituted by "n" parameters, generating a
fitness function. For this study, an identification of the rules and their
relationship with the respective variables was conducted first, allowing to
identify the best dredging for the channel (MENDELL et al, 1992; 1995).
In
population evolution, individual fitness was assessed first by the result of
the crossover operation and then of the mutation operation. Expert knowledge is
needed for the evolutionary computation in order to perform the deductive
reasoning. That is, an expert's knowledge is important when we wish to deduct
or infer a conclusion given a set of facts and knowledge. Many formulations
applicable to this study with respect to formal knowledge are presented in the
literature (ROSS, 1980).
5. RESULTS AND DISCUSSION
Learning in the ANFIS consisted in adjusting its
parameters, in two steps. The first one included the estimation of parameters,
which were assumed to be linear for this model. The second step involved a
process of parameter optimization through the application of rules defined for
the model according to a backpropagation method. During the optimization step,
the algorithm was made with the method of Conjugate Gradient (CG), a method
created to solve iterative linear problems. The solution obtained with this
method started from the hypothesis that symmetric coefficient matrices are
positivelydefined, and the method then progresses and converges in a finite
number of iterations.
The
definition of initial conditions considered the series of input data obtained
and tabulated to calculate flow rate, with a critical value corresponding to
the desired degree of confidence, of 0.8;
standard deviation of the flow rate variable of 0.5; and margin of
error of 0.1. This allowed the
definition of sample size as 16 (sixteen) observations, for triangular fuzzy
numbers (TRIOLA, F, 1999).
Training of the neuro-fuzzy network
in this study consisted in an exercise of numerical optimization of a nonlinear
function. The technical literature describes a number of methods of nonlinear
optimization (e.g. Bertsekas, 1995). In the present case, the option was to
obtain a solution with the aid of the "Solver®” software, which
solves this type of problem with a specific process. The use of the
"Solver®" software was considered convenient since it is
available on MS Excel®, where a numerical optimization add-in can be used to
expand Excel's capabilities, allowing the solution to be obtained in a quick,
easy and precise way (CHOONG, 2009).
The
ANFIS model considered a sample with a moderate number of weights. The data
were treated in an ANFIS in the "Solver"® software (Figure 2).
Technical literature suggests the method of Conjugated Gradients (CG) to
estimate the parameters. However, if parameters are interpreted with the local
data, parameter estimation is considered an important aspect for the RMS result,
and it may be convenient to use another method in which deviations can be
minimized around the line (or in this case, around the estimated hyperplane)
without biases (CHOONG, 2009).
Figure 2: Learning result of the
neuro-fuzzy network using the Solver® software in MS Excel®
After 18 (eighteen) observations,
"series 2", representing responses provided by the learning of the
trained ANFIS, had good superposition with "series 1" (experimental
data), indicating good learning by the neuro-fuzzy network. The model indicated
the flow rate activation function as being given by the following expression:
Y1
= 0,2- 0,1 . X1 + 0,8 . X2 + 0,1. X3 + 0,2.X4
(4)
Where Y1 is the flow rate in m³/s, X1 is the
channel bathymetry, X2 is the tide in the channel, X3is
the winds along the channel and X4 is the roughness along the
channel.
The
computer software used for simulating sediment transport is the same one used
for the hydrodynamic modeling, which will be presented below just for a
comparative analysis of results. Programming of the sediment transport model
used the results obtained from the hydrodynamic model (ANFIS). The model's
evolution was made from water levels collected previously and other
hydrodynamic data. The main parameters used in the mathematical modeling were:
modeling area: 32.5 km x 45.0 km; channel "Flow rate” forcing; and
"channel “Area" forcing.
This
study considered a fixed crossover rate, PC (Yi (x)), equal to 60%, and a value for
mutation rate PM (Yj (x)) equal to 1% mutation, as suggested by RODRIGUES F.L. et al (2004). Thus, the adjusted fitness
function for sediment transport is as follows:
Y2
= 0,9 . X1 + 0,1 . X2 – 1,85 (5)
where Y2 is the sediment concentration in
the channel, X1 = flow rate along the channel and X2 is
the channel's Area. Training of the GAF model had a RMS of 85% and a good
result for the model's statistical fit. Only the variable "Area" had
low significance level. This result, however, can be considered acceptable due
to the estimator's consistency analysis.
6. COMPARISON OF THE RESULTS OBTAINED WITH THE HYBRID
MODEL TO THOSE OBTAINED WITH THE FINITE ELEMENT MODEL(MIKE 21®)
Hydrodynamics and sediment transport
studies may be performed through simulations involving software using the
technique of finite elements, such as MIKE 21®. This section
involves the comparison of two sediment transport models. The first one is the
finite element-based model MIKE21®, provided by DHI - Danish
Hydraulic Institute and the second one is the model built in this study.
In
order to study the response of the proposed channel, initially, from the
hydrodynamic simulation developed using the hydrodynamic model (ANFIS), 3
(three) scenarios were proposed simulating the width and depth of the channel,
as described below:
Table 2: Scenarios studied for dredging of the “Linguado” channel / SC- Brazil
Scenarios |
Landfill
removal (m2) |
Landfill
removal (m2) |
Landfill
removal (m2) |
|||
North |
South |
North |
South |
North |
South |
|
A |
X (100,150,200,250) |
- |
- |
- |
- |
- |
B |
- |
- |
- |
X (100,150,200,250) |
- |
- |
C |
- |
- |
- |
- |
X
(100,150,200,250) |
X
(100,150,200,250) |
Source: IME, 2003.
These
scenarios involved hypothetical situations with different conditions for the
variables considered. Therefore, setting up the scenarios considered boundary
conditions that conducted to the definition of the variation fields for the
variables considered as being the same, with the goal of making these estimates
consistent for the purpose of assessing results for the whole set.
The process of solving
of the proposed model includes the definition of scenarios believed to have
more consistent results, which is an important aspect for decision making
processes. The results are summarized in the figures below and show that the
results obtained by the hybrid model have good superposition with those of the
model of MIKE21® software. Each figure represents a scenario, where
"series 2", in red, are the results obtained through the finite
element simulation (IME, 2003), while "series 1", in blue, shows the
results for the fuzzy-genetic model.
Figure 3: Evolution of
experiments for scenario "A"
Figure 4: Evolution of
experiments for scenario "B"
Figure 5: Evolution of
experiments for scenario "C"
The results indicate
that there is a superposition of the output of both models for scenarios
"A" and "B", which is expected due to the correlation of
0.98 between both graphics. This does not occur for scenario "C",
however, which considers dredging in both sides of the “Linguado” channel; in this case, the models have a superposition of
0.75. It may be considered, however, that the behavior of both models does not
show relevant distortions, in light of the aspects discussed above regarding
the data, allowing the conclusion that both models provided a good
representation of the phenomenon studied.
Sediment
transport values during a flow tide found maximum rates for scenario
"A", both for the MIKE21® simulation and for the one using
the fuzzy genetic algorithm. Simulations with both methods did not show
relevant differences, and the maximum flow rate value for this scenario was
1.01 cm/s. Some difference in values was observed for scenario "C",
likely because the historical time series for the "South" channel are
less reliable, considering that this channel has been closed for a long time
and all data used were theoretical extrapolations.
Sediment
transport gradients reduce simultaneously with increases in channel depth. This
indicates that the system with deeper channels is closer to equilibrium than
the system with shallower channels. The GAF model also allows concluding about
transported material. According to the Hybrid model, a flow rate of 1.01 cm/s
in scenario "A" corresponds to an NTU of 10, that is, a high level of
sediment and organic matter transport.
7. CONCLUSIONS
This study involved a
review of theories for modeling hydrodynamic and sedimentological processes and
a review of fuzzy logic and other hybrid systems. The analysis of the various
possible theoretical modeling perspectives allowed the identification of the
most convenient steps to model the process studied here.
Scenarios
were defined based on analyzing their relevance for the stability analysis of a
tidal channel. The “Linguado” Channel
in the landfill area is not strictly a tidal channel, although the aspects
related to physical mechanisms determining its stability are similar. In this
manner, the model used here allows the study of sediment-derived processes, including
erosion, sediment transport in water bodies and sediment deposition.
During the search for a better understanding of
abstract and concrete aspects involved in the studied scenario and their
influences in modeling, a review of the theoretical references relevant to the
scenario presented has been conducted, assessing how they could have influenced
the decisions in process modeling works, from the closure of the “Linguado” Channel in 1935 to the
possible implications of this closure, since it may have caused the deposition
of a large amount of sediments in the area surrounding the landfill and
especially in its northern side, where considerable silting has been happening.
This closure may also have caused a general reduction in channel depths in
comparison to the period before its closure.
The
hybrid models used (neuro fuzzy and fuzzy genetic) were able to learn and
reproduce the studied phenomenon; predict the “Linguado” Channel's flow rate, which is significantly influenced by
the tide in Babitonga Bay; and also identify the strong correlation which
exists between the “Linguado”
Channel's flow rate and the total amount of sediments transported by the
Channel.
This
study established that the sediment input coming from Babitonga Bay is
considerable, being the largest contributor to sediments in the ““Linguado”” Channel region in “Santa Catarina”. Additionally, it
established that, despite the considerable sediment input coming from “Babitonga” Bay, depth variations in the
“Linguado” Channel occurred differently
from elsewhere in the region.
REFERENCES
ASSAD, L. P. F. et al. (2009). Basic
course on computational hydrodynamic modeling and dispersion of pollutants, May
2009, Laboratory of Computational Methods -
UFRJ.
BRAIT,
M.; DALZOTTO, D. (2010). Numeric Simulations
Shall Waters, CNMAC, p.1026.
CHOONG, J. (2009). Build Neural Network With MS Excel ®,
Published by XLPert Enterprise Copyright © 2009 by XLPert Enterprise.
FEMAR, (2000). Catalog of Brazilian Maregraphic Stations . Fundação
de Estudos do Mar. Rio de Janeiro, p.281, RJ.
GRACEA, V. B.; MAS-PLAC, J.; NOVAISB, T. O.;
SACCHID, E.; ZUPPIA, G. M. (2008) .Hydrological
mixing and geochemical processes characterization in an estuarine/mangrove
system using environmental tracers in Babitonga Bay (Santa Catarina, Brazil),
Vol. 28 , p.682 – 695.
HERKENHOFF, L.; FOGLI, J. (2013). Applied Statistics for Business and
Management using Microsoft Excel, © Springer Science, New York,
ISBN 978-1-4614-8422-6.
IME
& UNIVALE. (2003). Diagnóstico dos estudos de
circulação de água no canal do “Linguado” e na Baía de Babitonga/SC.
JANG, J. S. R.
(1993). ANFIS: Adaptive – Network - Based Fuzzy Inference System. IEEE
TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS, v. 23, n. 3, May / Jun.
JANG, J. S. R.; SUN, C. T.; MIZUTANI, E. (1997). Neuro-Fuzzy and Soft
computing – A computational approaching learning and machine intelligence, Prentice Hall.
KLEMENS A. L. A.
(2012). Environmental
legislation with the focus of conflicts: an analysis from the social
representations of the nature of small farmers in Minas Gerais, Brasil, Soc. & Nat., Uberlândia, v. 24, n. 3,
p. 405-418, Sept / Dec.
REZENDE S. O. et al. (2005). Intelligent Systems -
Fundamentals and Applications.
Manole ltd.
RODRIGUES A. M.
T. (2000). Socio-Economic
Diagnosis and the environmental perception of communities of artisanal
fishermen from the Bay of Babitonga: a subsidy to coastal management. UFSC / Environment Department.
ROSS, T. J. (2004). Fuzzy logic
with Engineering application, 2nd edition, John & Willey Sons
Ltd.
TRIOLA, M. F.
(1999). Introduction
to Statistics. LTC. 7th. ed. Rio de Janeiro.
VIEIRA C. V. et al.
(2008). Morphosedimentary characterization and sectorization of the estuary
complex of Babitonga Bay. Geosciences Paranaense Publication, UFPR, v. 62-63, p. 85-105.
RUSSEL S.; NORVIG P. (2008). Artificial
intelligence, a modern approach. Pearson Prentice Hall, 2nd ed., Madrid.